Rodríguez-Delgado et al. BMC Evolutionary Biology (2014) 14:267 DOI 10.1186/s12862-014-0267-z RESEARCH ARTICLE Open Access Paternal X inactivation does not correlate with X chromosome evolutionary strata in marsupials Claudia L Rodríguez-Delgado1*, Shafagh A Waters2 and Paul D Waters2* Abstract Background: X chromosome inactivation is the transcriptional silencing of one X chromosome in the somatic cells of female mammals. In eutherian mammals (e.g. humans) one of the two X chromosomes is randomly chosen for silencing, with about 15% (usually in younger evolutionary strata of the X chromosome) of genes escaping this silencing. In contrast, in the distantly related marsupial mammals the paternally derived X is silenced, although not as completely as the eutherian X. A chromosome wide examination of X inactivation, using RNA-seq, was recently undertaken in grey short-tailed opossum (Monodelphis domestica) brain and extraembryonic tissues. However, no such study has been conduced in Australian marsupials, which diverged from their American cousins ~80 million years ago, leaving a large gap in our understanding of marsupial X inactivation. Results: We used RNA-seq data from blood or liver of a family (mother, father and daughter) of tammar wallabies (Macropus eugenii), which in conjunction with available genome sequence from the mother and father, permitted genotyping of 42 expressed heterozygous SNPs on the daughter’s X. These 42 SNPs represented 34 X loci, of which 68% (23 of the 34) were confirmed as inactivated on the paternally derived X in the daughter’sliver; the remaining 11 X loci escaped inactivation. Seven of the wallaby loci sampled were part of the old X evolutionary stratum, of which three escaped inactivation. Three loci were classified as part of the newer X stratum, of which two escaped inactivation. A meta-analysis of previously published opossum X inactivation data revealed that 5 of 52 genes in the old X stratum escaped inactivation. Conclusions: We demonstrate that chromosome wide inactivation of the paternal X is common to an Australian marsupial representative, but that there is more escape from inactivation than reported for opossum (32% v 14%). We also provide evidence that, unlike the human X chromosome, the location of loci within the oldest evolutionary stratum on the marsupial X does not correlate with their probability of escape from inactivation. Keywords: X inactivation, Marsupial, Imprinting, X chromosome, Escape, RNA-seq, SNP, Evolution, Strata Background single X in males were upregulated [3]. Carry through of Therian (i.e. eutherian and marsupial) mammals have a X chromosome upregulation to females would result in XX female/XY male sex chromosome system, or some functional X tetrasomy (i.e. 4X: 2A gene dosage), which simple variant of it. The X and Y chromosomes evolved was subsequently balanced by transcriptional silencing of from an ordinary pair of autosomes via a process of gene one X in the somatic cells of females – called X chromo- loss on the Y [1], which resulted in an imbalance of X some inactivation (XCI) [4]. gene dosage with the autosomes (1X: 2A gene dosage) in During early embryogenesis, in eutherian females, the males. It was proposed that to rebalance transcriptional X inactive specific transcript (XIST)isexpressedfromthe output between the autosomes and the X chromosome, X inactivation center (Xic) on the future inactive X genes (or a subset of dosage sensitive genes [2]) on the chromosome (Xi). XIST, a long non-coding RNA (lncRNA), coats the X chromosome in cis and triggers a cascade of * Correspondence: [email protected]; [email protected] epigenetic events that stably represses transcription of 1Division of Evolution, Ecology and Genetics, Research School of Biology, The most X loci [5] (reviewed in [6]). In human, genes in older Australian National University, Canberra, ACT 0200, Australia regions of the X chromosome are more likely to be 2School of Biotechnology and Biomolecular Sciences, Faculty of Science, University of New South Wales, Sydney, NSW, 2052, Australia © 2014 Rodríguez-Delgado et al.; licensee BioMed Central. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Rodríguez-Delgado et al. BMC Evolutionary Biology (2014) 14:267 Page 2 of 8 inactivated, whereas genes located in the evolutionarily on evidence from four genes in six different marsupial younger regions of the X chromosome are more likely species [17]. Consistent with this, a more recent study to be expressed from the Xi (the escaper genes) [7,8]. The [34] confirmed preferential paternal inactivation as a choice of which X chromosome (maternal or paternal) be- chromosome-wide phenomenon in opossum brain and comes inactivate is random [4,9-12]. Although some prop- extraembryonic tissues. Here we analysed expression of X- erties of XCI mechanism vary between species (like timing linked genes in liver and blood from an Australian marsu- of XCI initiation [12]), the general epigenetic profile of the pial, the tammar wallaby. We sequenced polyadenylated Xi appears to be well conserved across eutherian mam- RNA (RNA-sequencing) from a father and mother blood mals [9,13-15]. sample, along with a liver sample from the daughter. To Eutherians diverged from the marsupial lineage ~180 overcome the limitation of some alleles not being present million years ago (MYA). The marsupial X is homolo- in the mother transcriptome due to silencing via XCI, we gous to the long arm and proximal short arm of the hu- also used DNA-sequencing data available from the father man X (denoted as the X conserved region XCR [16]). and the mother [36]. Like eutherian mammals, one of the two Xs in marsu- pials is inactivated [17], and although some epigenetic Mapping results and SNP calling features of the Xi are conserved between clades [18], An average of ~50 million (M) reads were obtained for there are also considerable differences [19,20]. The re- each of the three samples, of which approximately 26 M, gion homologous to the eutherian Xic has been dis- 20 M and 30 M were uniquely mapped for father, rupted on the marsupial X chromosome [21-23] and mother and daughter respectively (Additional file 1). In there is no XIST homologue [24]. Instead, the lncRNA total 44,095 SNPs were called between the wallaby refer- gene RSX (RNA on the silent X) appears to be involved ence genome and our three samples. 68% of SNPs were in marsupials XCI [25]. Like XIST, the RSX transcript is shared between the three individuals (due to being a exclusively expressed from the Xi and coats it in cis [25]. captive bred colony), so were non-informative when de- Early studies of allozyme variants and somatic cell hy- termining parent-of-origin of SNPs in the daughter. A brids of four genes in different marsupial species (G6PD, total of 356 SNPs (~1.5% of total) called (in at least one PGK1, GLA and HPRT [26-30]) proposed that marsupial sample) against the reference genome were assigned to XCI was incomplete and variable between tissues and spe- the X chromosome (Additional file 2). SNPs that were as- cies (reviewed in [17]). Additionally, the results showed sociated with genes (or scaffolds with genes) that did not that there was imprinted inactivation of the paternally have a 1:1 orthologue in human and/or opossum, plus derived X chromosome (as has been described for ex- those that we were unable to anchor to the X chromo- traembryonic tissues of rodents and cow [31-33]). A recent some, were not included in the analysis (see Methods). genome-wide study in opossum (Monodelphis domestica) Some reported SNPs were located outside wallaby anno- fetal brain and extraembryonic tissues reported complete tated genes and/or exons, likely due to the poor assembly paternal inactivation for most (86%) genes on the X and annotation of the tammar wallaby genome, and were chromosome, and escape from XCI for 14% of genes [34]. still considered in the analysis. This level of escape is comparable to that reported for the whole human X (15%). However, it is higher than Assessing allele-specific expression the level of escape on both the mouse X (3%) [35], and We calculated relative expression of each allele for het- the regions of the human X homologous to the marsupial erozygous biallelically expressed SNPs in the daughter’s X (6.5%) [7,35]. liver and the mother’s blood (as performed by Wang Comparison of XCI between different tissues and spe- et al. [34]). In addition, we also included SNPs that were cies offers new insights to its mechanism and evolution. confidently called as heterozygous in daughter (following In this study we used RNA-Seq data from a captive bred Mendelian inheritance based on parental genotype), but tammar wallaby family (i.e father, mother and daughter), that were monoallelically expressed (i.e. SNPs that ap- and genome sequencing from the father and the mother peared as homozygous in the transcriptome). As such, [36], to determine the parent of origin of expressed alleles we could calculate relative expression of each allele for on the daughter’s X chromosome. We confirm paternal both biallelically and monoallelically expressed heterozy- inactivation on the tammar wallaby X chromosome, but gous SNPs. observed more escape from inactivation than previously We observed apparent biallelic expression from the X reported for human, mouse and opossum. chromosome in the father, likely because Y-derived reads from the male were reported as best mapped to the X Results and discussion chromosome.